The good news is that this approximation is carried out with temporal coherence.

好消息是，这种近似是在时间相干的情况下计算的。

The limitations of this technique includes the inability of detecting expressions that are very far away from the ones seen in the training set, and as you can see in the videos, temporal coherence could also use some help.

3D技术的局限性包括，不能检测那些人们在训练装置中看到的偏差很大的表情或者你可以在录像中看到的表情，片段的连贯性也需要改进。